© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 2723
    Prev
    Next
    INDEX
    Explanations

    references to natural disasters, particularly floods

    oai_token-act-pair · gpt-3.5-turbo

    references to flooding events and their impacts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Notes
    -0.81
    CTV
    -0.74
     Vance
    -0.67
     Accessories
    -0.66
    Mart
    -0.66
    iem
    -0.65
    Dex
    -0.65
     Vert
    -0.65
    notes
    -0.65
    irmed
    -0.64
    POSITIVE LOGITS
     flood
    3.85
     Flood
    2.41
     floods
    2.30
     flooding
    2.28
     flooded
    2.11
     inund
    1.85
     tsunami
    1.74
     tidal
    1.60
     hurricane
    1.45
     earthquake
    1.44
    Activations Density 0.014%

    No Known Activations